Create a Message changedapi/beta/messages/create
Nearest release: v2.1.247, published under an hour before this site recorded the change. Shown because the two are within 24 hours of each other. Nothing here says the release caused the edit.
Recorded here
Lines+29added
Lines−7removed
From line
16
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits30to this page, all time
The whole hunk
from line 16, old and new numbered
/
from line 16
1616
1717 - `string`
1818
19 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 31 more`
19 - `"message-batches-2024-09-24" or "prompt-caching-2024-07-31" or "computer-use-2024-10-22" or 38 more`
2020
2121 - `"message-batches-2024-09-24"`
2222
from line 86
8686
8787 - `"mid-conversation-tool-changes-2026-07-01"`
8888
89 - `"compact-2026-01-12"`
90
91 - `"computer-use-2025-11-24"`
92
93 - `"mcp-tunnels-2026-06-22"`
94
95 - `"structured-outputs-2025-11-13"`
96
97 - `"task-budgets-2026-03-13"`
98
99 - `"thinking-display-updates-2026-08-18"`
100
101 - `"ce-user-management-2026-07-13"`
102
89103- `"anthropic-user-profile-id": optional string`
90104
91105 The user profile ID to attribute this request to. Use when acting on behalf of a party other than your organization. Requires the `user-profiles` beta header.
from line 1780
17661780
17671781 - `type: "enabled"`
17681782
1769 - `display: optional "summarized" or "omitted" or null`
1783 - `display: optional "summarized" or "omitted" or "updates" or null`
17701784
17711785 Controls how thinking content appears in the response. When set to `summarized`, thinking is returned normally. When set to `omitted`, thinking content is redacted but a signature is returned for multi-turn continuity. Defaults to `summarized`.
17721786
from line 1788
17741788
17751789 - `"omitted"`
17761790
1791 - `"updates"`
1792
17771793 - `BetaThinkingConfigDisabled object`
17781794
17791795 - `type: "disabled"`
from line 1798
17821798
17831799 - `type: "adaptive"`
17841800
1785 - `display: optional "summarized" or "omitted" or null`
1801 - `display: optional "summarized" or "omitted" or "updates" or null`
17861802
17871803 Controls how thinking content appears in the response. When set to `summarized`, thinking is returned normally. When set to `omitted`, thinking content is redacted but a signature is returned for multi-turn continuity. Defaults to `summarized`.
17881804
from line 1806
17901806
17911807 - `"omitted"`
17921808
1809 - `"updates"`
1810
17931811 - `Default = "default"`
17941812
17951813- `inference_geo: optional string or null`
from line 5318
53005318
53015319 Per-iteration token usage breakdown.
53025320
5303 Each entry represents one sampling iteration, with its own input/output token counts and cache statistics. This allows you to:
5321 Each entry represents one sampling iteration, with its own input/output token counts and cache statistics, discriminated by `type`. For `message` entries (model sampling iterations, such as the turns of a server-side tool use loop), this allows you to:
53045322
53055323 - Determine which iterations exceeded long context thresholds (>=200k tokens)
5306 - Calculate the true context window size from the last iteration
5324 - Calculate the context window size from the last `message` entry
53075325 - Understand token accumulation across server-side tool use loops
53085326
5327 A `compaction` entry reports the token usage of the compaction operation itself — the server-side request that summarizes the context being closed — NOT the size of the context that was compacted away, and its token counts can be much smaller than that closed context (for example, a compaction that closes a ~200k-token context can report only a few thousand tokens). Do not derive the context window size from a `compaction` entry, even when it is the last entry. A `compaction` entry's tokens are not included in the top-level `usage` fields. When an input-token trigger is in effect (the default — 150,000 tokens unless configured otherwise), each `compaction` entry closes a context that had reached at least that threshold, though the context can exceed it by the final iteration's output and tool results.
5328
53095329 - `BetaMessageIterationUsage object`
53105330
53115331 Token usage for a sampling iteration.
from line 5635
56155635
56165636 Per-iteration token usage breakdown.
56175637
5618 Each entry represents one sampling iteration, with its own input/output token counts and cache statistics. This allows you to:
5638 Each entry represents one sampling iteration, with its own input/output token counts and cache statistics, discriminated by `type`. For `message` entries (model sampling iterations, such as the turns of a server-side tool use loop), this allows you to:
56195639
56205640 - Determine which iterations exceeded long context thresholds (>=200k tokens)
5621 - Calculate the true context window size from the last iteration
5641 - Calculate the context window size from the last `message` entry
56225642 - Understand token accumulation across server-side tool use loops
5643
5644 A `compaction` entry reports the token usage of the compaction operation itself — the server-side request that summarizes the context being closed — NOT the size of the context that was compacted away, and its token counts can be much smaller than that closed context (for example, a compaction that closes a ~200k-token context can report only a few thousand tokens). Do not derive the context window size from a `compaction` entry, even when it is the last entry. A `compaction` entry's tokens are not included in the top-level `usage` fields. When an input-token trigger is in effect (the default — 150,000 tokens unless configured otherwise), each `compaction` entry closes a context that had reached at least that threshold, though the context can exceed it by the final iteration's output and tool results.
56235645
56245646 - `output_tokens: number`
56255647
No line in this hunk matches that.